Part of the DoW SBIR/STTR FY27 Release 2 topic list. Browse all 83 topics by industry or service.
Topic DPA27BZ02-DV003 from the DoW SBIR 2027 BAA, Release 2. Pre-release now; proposals due January 6, 2027, 12:00 PM ET.
Quick answer
DPA27BZ02-DV003 is a DARPA SBIR topic seeking a non-invasive wearable muscle-sensing interface (sEMG, PMUTs, or similar) that lets a warfighter control drones, radios, and TAK with gestures while keeping hands and eyes on the primary task. It is Direct-to-Phase II only and requires an existing working prototype. Proposals are due January 6, 2027, 12:00 PM ET.
Solicitation at a glance
- Topic number: DPA27BZ02-DV003
- Official title: Human Universal Remote
- Agency: DARPA
- Program: SBIR (DoW SBIR 2027 BAA, Release 2)
- Award type: Direct to Phase II (no Phase I awards)
- CMMC level: Level 2 (Self)
- ITAR: Yes. Export-controlled (ITAR/EAR); see topic text
- Technology areas: Electronics, Human Systems, Sensors
- Pre-release (direct TPOC contact allowed): November 4 to December 2, 2026, 8:00 AM ET
- Opens in DSIP: December 2, 2026, 8:00 AM ET
- Proposals due: January 6, 2027, 12:00 PM ET
- Official source: DSIP Topics app (search DPA27BZ02-DV003)
BW&CO's take
What the DARPA wants
DARPA wants a hands-free, eyes-free human-machine interface built from forearm-conforming muscle sensor arrays and low-latency gesture recognition. TA1 covers signal capture that survives sweat, movement, collisions, and EMI while also monitoring fatigue and stress. TA2 covers translating intentional micro- and macro-gestures into standardized commands sent over at least one RF channel, built for TAK integration. Both TAs must be addressed together. Phase II runs 18 months or less, with readiness tests at Month 12 and Month 18 requiring 92% true-positive gesture recognition, 2% false alarms, and calibration under 30 minutes, the second in an austere field setting.
Who is a good fit
- A working high-density sEMG electrode array with real-time machine-learning gesture classification already demonstrated
- Existing calibration and training software that measures classification accuracy and misclassification rates
- Ruggedized wearable hardware experience, including sweat, motion, EMI resistance, and sealing toward an IP68 rating
- Integration experience with ATAK, Bluetooth or other RF links, radios, or unmanned system control
Watch-outs
- Direct-to-Phase II only: the proposal must already contain evidence of a working prototype, including a non-invasive sEMG array, real-time gesture classification hardware, and a calibration software interface.
- Both technical areas (signal capture and command output) must be addressed concurrently in one integrated solution, and multi-platform validation must cover two of: ATAK control, communication device control, communication transmission, or unmanned system tasking.
- The readiness tests use previously untrained subjects, require calibration to adapt without developer intervention, add high-exertion noise, and the Month 18 field test includes CBRN protective gear, collision testing, and simulated weapon manipulation.
- The topic metadata marks it as ITAR-restricted, so plan for export-control handling of technical data and possible limits on foreign national participation. The metadata lists CMMC Level 2 (Self).
Think your company fits DPA27BZ02-DV003? Talk with BW&CO before the January 6 deadline.
Plan Your DPA27BZ02-DV003 Submission
Official topic text (verbatim from DSIP)
The text in this section is copied word for word from the official DSIP topic as published on November 4, 2026. DSIP is the official source and topics can be amended. Always confirm the current version at dodsbirsttr.mil before you submit. Our commentary above is BW&CO's interpretation, not Government guidance.
Title
Human Universal Remote
Objective
The objective of the Human Universal Remote DP2 SBIR topic is to further develop surface electromyography (sEMG) sensors, piezoelectric micromachined ultrasonic transducers (PMUTs), or other non-invasive sensor modalities as a user interface for warfighter multi-system operations1-6. The cited references serve as representative baselines of the state-of-the-art human machine interfaces and do not limit the scope of acceptable technical approaches.
The vision is for a single warfighter to execute simultaneous, multi-domain tasks without compromising their primary physical posture or situational awareness. For example, controlling a drone while handling a rifle, or sending/receiving audible, out of line-of-sight communications via radio without speaking. The envisioned system may combine micro-scale ultrasonic cellular stimulation, ultrasound and EMG sensor fusion, and deep learning driven gesture recognition. The system must translate and discern between sub-millimeter muscle contractions and intentional, full-range movements to trigger external system commands. In a series of readiness tests, performers will develop the sensor suite and processing framework to serve as a hands-free, eyes-free human-machine interface for multi-system warfighter operations.
Description
Surface electromyography (sEMG) is a proven technique for detecting biomedical voltage measurements from the skin generated by the electrical current of muscle fibers during skeletal muscle contraction and relaxation. These signals, which reflect neuromuscular activity, typically range in amplitude from 0 to 10 millivolts and operate at frequencies between 20 and 450 Hz. While historically used to restore hand function for transradial amputees via non-invasive fiber membrane potential measurement, this topic proposes leveraging state-of-the-art sEMG in concert with other non-invasive muscle sensing modalities to significantly enhance the operational capability of the warfighter. For example, ultrasound is a high-resolution non-invasive sensing modality capable of probing deep muscle tissues and providing complementary information to electromyographic signals. This topic is divided into two distinct, parallel Technical Areas (TAs) that must be addressed concurrently. Proposers must deliver an integrated solution spanning both domains:
Technical Area 1 (TA1): Microphone
Signal capture will focus on the non-invasive detection and extraction of EMG signals and/or electrical, acoustic deep muscle sensing signals from the user. The approach should include the utilization of signals for arm control as both a command and as biological sensors to understand where muscle activity is occurring and to function as an active indicator of warfighter stress. Hardware systems developed under this Technical Area must reliably detect, isolate, and extract these microvolt-to-millivolt level biopotential signals directly from the warfighter’s skin without invasive techniques. Proposers should address the following TA1 criteria:
TA1 Criteria | Description |
Array design | Single or multimodal sensor arrays that conform to the forearm or targeted muscle groups |
Form factor | Integrates with standard-issue warfighter apparel without causing skin irritation or impeding range of motion over extended wear |
Ruggedization | Maintain a high signal-to-noise ratio under stress:
|
Dual-use sensing | Capture intent-driven neuromuscular firing patterns to facilitate hands-free control of external systems while serving as passive biological monitoring of muscle fatigue, physical overexertion, and cognitive stress |
Calibration | The system must be designed for recalibration once during initial outfitting of the sensor array onto the warfighter. Calibration techniques should require minimal to no additional cognitive load and take as much time as to be doable during mission briefing. |
Technical Area 2 (TA2): Speaker
This Technical Area focuses on translating decoded physiological intent into standardized, platform-agnostic output modalities. The system must reliably differentiate between passive muscle strain and active, intentional micro- and macro-gestures to generate high-confidence external commands. The approach should include command output that translates the processed signals into an actionable trigger via standard communications channels designed for easy integration into Team Awareness Kit (TAK). To prevent over-engineering to a singular use case or platform, the generated signaling protocol should be standardizable for use across multiple external systems. At a minimum, the platform should be built with at least one communications channel using radiofrequency (RF). Example platforms include enabling users to task robotic systems, low-level ISR drones, comms drones, trigger communications, and generate visual indicators directly on a TAK display. Proposers should address the following TA2 criteria:
| TA2 Criteria | Description |
| Compute | Implement low latency machine learning or pattern-recognition algorithms capable of running on power-constrained processors |
| Modular Output | Generate control packets transmittable via RF waveforms such as Bluetooth |
| Multi-Platform | Validate the versatility of decoded gesture triggers (pick 2):
|
Phase I
Since this is a DP2 topic ONLY, a successful proposal must already contain evidence of a working prototype. The success criteria for a working prototype are the following:
- Working non-invasive sEMG sensor array that employs a computationally efficient machine-learning based control algorithm driven by input from a high-density array of electrodes.
- Functioning hardware that interfaces with a machine learning algorithm capable of real-time gesture classification.
- A developed training and calibration software interface that prompts users through gesture protocols, records performance metrics, measures gesture classification accuracy, and misclassification error rates.
The above information should be presented in such a manner that it is evident that the proposer’s prototype technology can be tailored to meet the criteria in Phase II.
Phase II
A successful proposal will address the overarching technical areas with performer-defined metrics over the course of 18 months or less. The program is structured with readiness tests at Month 12 and Month 18. Proposers will be expected to define at least one deliverable per month to meet incremental technical milestones to sufficiently advance the technology to meet the Controlled Environment Readiness and Environmental Resilience tests. Proposals should outline a work plan to technically advance the design to meet the demonstration scenarios within the provided timeline.
Controlled Environment Readiness Test – Jam Session
This readiness test will serve as the gatekeeper metric to prove the system can successfully transition from controlled development environments into unpredictable operational settings. The system must be tested on a diverse cohort of previously untrained and unexposed human subjects. The software calibration must adapt to varied anatomical structures, body mass indexes, and physiological signal variations without developer intervention. Testing will occur during a series of continuous baseline of physical activities. These activities will include high-exertion tasks designed to introduce physical noise to test rejection of background muscle activity. While performing the baseline physical activities, subjects will be prompted to execute specific micro- and macro-command gestures. The system must dynamically isolate and decode these intentional commands from the background physiological "noise" of the active baseline tasks.
| Session Criteria | Description |
| Gesture accuracy | 92% true-positive command recognition across the user pool |
| False alarm | 2% accidental command triggers during continuous baseline physical activities |
| Calibration | Calibration and training in under 30 minutes |
Environmental Resilience – Arena Concert
The Month 18 evaluation will transition the system from controlled laboratory settings into an austere field environment at a representative military training installation (e.g., Marine Corps Base Camp Pendleton), subjecting the user to the following parameters while executing gestures for one command output that translates the processed signals into an actionable trigger via RF and/or ATAK integrations:
- Ingress Protection (IP) rating IP68
- In tandem with standard military uniform/dress
- Chemical, biological, radiological, and nuclear (CBRN) protective gear
- High-exertion physical activity/collision testing
- Simulated weapon manipulation
Gesture Output can include any of the following, for example:
- Audible communication without speaking and out of line-of-sight
o Ground and Water Testing
- Autonomous system tasking of any variety
- TAK manipulation
- Radio control
- Demonstration of ability to remove radio from weapons space
| Concert Criteria | Description |
| Gesture accuracy | 92% true-positive command recognition across the user pool |
| False alarm | 2% accidental command triggers during physical activities |
| Calibration | Calibration and training in under 30 minutes |
Phase III Dual Use Applications
Phase III will be oriented toward military transition and further commercialization.
- Military: Autonomous system tasking, enabling raid teams to communicate without speaking, and providing visuals and status through fatal funnel.
- Commercial: Accessibility technology, quantitative biomechanical baselines to monitor neuromuscular health for rehabilitation, open-source code for personal drone piloting and gaming applications.
References
Balasubramanian, P. S., Singh, A., Xu, C., & Lal, A. (2020). GHz ultrasonic chip-scale device induces ion channel stimulation in human neural cells. Scientific Reports, 10(1), Article 3075.
Rabe, K. G., & Fey, N. P. (2022). Evaluating electromyography and sonomyography sensor fusion to estimate lower-limb kinematics using Gaussian process regression. Frontiers in Robotics and AI, 9, Article 716545.
Sharma, V., Lloyd, E., Faltys, M., Ortiz-Catalan, M., & Glass, C. (2025). Assessment of gesture accuracy for a multi-electrode EMG-sensor-array-based prosthesis control system. Prosthesis, 7(4), 99.
Scheme, E. J., Englehart, K. B., & Hudgins, B. S. (2011). Selective classification for improved robustness of myoelectric control under nonideal conditions. IEEE Transactions on Biomedical Engineering, 58(6), 1698–1705.
Shi, X., He, Z., Zheng, L., Chen, Y., Bai, W., Xu, M., & Zhu, R. (2026). MSCB-DualAttn Network for sEMG-Based Gesture Recognition in Transradial Amputees with Varying Residual Limb Lengths. Symmetry, 18(5), 869.
Yang, W., Tang, Z., Zhu, K., Zhou, G., & Xie, J. (2026). Wearable multimodal sensing using piezoelectric micromachined ultrasonic transducer and surface electromyography for hand gesture recognition. Measurement, 290, Article 110650.
Keywords
Human Remote Control, Human Universal Remote, Autonomous System Tasking, Wearables, EMG, Human/Machine Teaming, Human/Machine Interface, Biosignal Translation
Frequently asked questions
What is DPA27BZ02-DV003?
It is a DARPA SBIR topic in the DoW SBIR 2027 BAA titled Human Universal Remote. It seeks non-invasive muscle sensing (sEMG, PMUTs, or other modalities) as a gesture interface for warfighter multi-system operations, such as controlling a drone while handling a rifle.
When is DPA27BZ02-DV003 due?
Proposals are due January 6, 2027, 12:00 PM ET. Pre-release began November 4, 2026, and direct contact with the topic authors (TPOCs) is allowed only until the topic opens on December 2, 2026, 8:00 AM ET.
Is this Phase I or Direct-to-Phase II?
Direct-to-Phase II only. Phase II runs 18 months or less, with readiness tests at Month 12 and Month 18, and proposers must define at least one deliverable per month.
What are the pass criteria for the readiness tests?
Both the Month 12 Controlled Environment Readiness Test and the Month 18 Environmental Resilience test require 92% true-positive command recognition across the user pool, 2% accidental command triggers during physical activity, and calibration and training in under 30 minutes. The Month 18 test adds IP68, military uniform, CBRN protective gear, collision testing, and simulated weapon manipulation at a representative military training installation.